What the Difficulty Tells You

If you find this exercise frustrating, you're not alone. Most people do. The frustration usually comes from one of two places. Either you're realizing that more of your job is ripe for automation than you thought, which is uncomfortable. Or you're realizing that you've been inflating the importance of tasks that are actually Tier 1, which is also uncomfortable. Either way, the discomfort is information.

If you notice you're gravitating toward "this is too nuanced for AI" as an explanation for why your work is safe, ask yourself: is that true, or is that the story I'm telling myself? If it's true, great. You should be able to point to specific things that require judgment in a way that's genuinely hard to automate. If you're struggling to articulate what those things are, you might be in denial mode.

If you notice you're catastrophizing and assuming everything you do is threatened, ask yourself: is that specific to your role, or am I applying the general capability of AI to my specific situation? AI can do a lot of Tier 1 work. But it can't do all of it, and it definitely can't do the Tier 2+ work that shapes and directs what it does.